INDEX

Explanations

elements and functionalities related to auto-completion in user interface components

oai_token-act-pair · gpt-4o-mini Triggered by @bot

autocompletion and item suggestions

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

exitRule

-0.47

UnusedPrivate

-0.44

 servlet

-0.40

AxisAlignment

-0.37

 newBuilder

-0.37



-0.36

 dateFormat

-0.35

 CreateTagHelper

-0.35

 Paglinawan

-0.35

chamber

-0.34

POSITIVE LOGITS

EndGlobalSection

0.55

 kasarigan

0.52

featureID

0.50

Géographie

0.49

pecabe

0.47

 colgante

0.43

oa̍t

0.42

 <<<<<<<<<<<<<<

0.42

$_['

0.42

 ujednoznacz

0.42

Activations Density 0.040%

elements and functionalities related to auto-completion in user interface components

autocompletion and item suggestions

No Comments

No Known Activations